Immobilized metal ion affinity chromatography (IMAC) was first introduced by Porath (Porath, J., J. Carlsson, I. Olsson, G. Belfrage 1975! Nature 258:598-599.) under the term metal chelate chromatography and has been previously reviewed in several articles (Porath, J. 1992! Protein Purification and Expression 3:263-281; and articles cited therein). The IMAC purification process is based on the employment of a chelating matrix loaded with soft metal ions such as Cu.sup.2+ and Ni.sup.2+. Electron-donating groups on the surface of proteins, especially the imidazole side chain of histidine, can bind to the non-coordinated sites of the loaded metal. The interaction between the electron donor group with the metal can be made reversible by lowering the pH or by displacement with imidazole. Thus, a protein possessing electron-donating groups such as histidine can be purified by reversible metal complex/protein interactions.
Several different metal chelating ligands have been employed in IMAC to purify proteins. Iminodiacetic acid (IDA) ligand is a tridentate and thus anchors the metal with only three coordination sites (Porath, J., B. Olin 1983! Biochemistry 22:1621-1630). Because of the weak anchoring of the metal, metal leakage has been known to occur. The tris(carboxymethyl)ethylenediamine (TED) ligand is pentadentate and forms a very strong metal-chelator complex. The disadvantage of this is that proteins are bound very weakly since only one valence is left for protein interaction. Nitrilo triacetic acid (NTA) is a tetradentate ligand which attempts to balance the metal anchoring strength with metal-ion protein interaction properties (Hochuli, E., H. Dobeli, A. Schacher 1987! J. Chromatography 411:177-184). Other chelating ligands have been reported and are mentioned. See, e.g. Porath (1992), supra. However, these ligands also have certain disadvantages, including decreased bonding capacity, decreased specificity, and increased metal leakage.
In 1991, Ford et al. (Ford, C., I. Suominen, C. Glatz 1991! Protein Expression and Purification 2:95-107) described protein purification using IMAC technology (Ni-NTA ligand) as applied to recombinant proteins having tails with histidine residues (polyhistidine recombinant proteins). This method takes advantage of the fact that two or more histidine residues can cooperate to form very strong metal ion complexes. The NTA chelating ligand immobilized on agarose and loaded with Ni.sup.2+ has been useful in this method (Hochuli et al., supra; U.S. Pat. No. 5,047,513). It is available commercially through Qiagen, Inc. (Chatsworth, Calif.). However, this resin has the disadvantage that the interchanges between metal ions and poly-histidine recombinant proteins are not optimal. Metal leakage can occur, and background proteins can sometimes contaminate purification of recombinant proteins.
A metal chelating gel, i.e., carboxymethylated aspartate (CM-Asp) agarose complexed with calcium, has been used for purifying native calcium-binding proteins (Mantovaara, T., H. Pertoft, J. Porath 1989! Biotechnology and Applied Biochemistry 11:564-570; Mantovaara, T., H. Pertoft, J. Porath 1991! Biotechnology and Applied Biochemistry 13:315-322; Mantovaara, T., H. Pertoft, J. Porath 1991! Biotechnology and Applied Biochemistry 13:120-126). However, the Ca.sup.2+ -CM-Asp complex described by Mantovaara et al. has among its disadvantages that it does not bind strongly to histidine-tagged recombinant proteins. Another disadvantage, in addition to this inferior binding property, is its non-selectivity for histidine tags.
By contrast, the subject invention comprises the CM-Asp chelating ligand complexed to a transition metal in an octahedral geometry (coordination number of 6). In this unique configuration, the metal complex can be advantageously suited for purification of poly-histidine fused recombinant proteins. This is a novel use of the CM-Asp ligand and is part of the subject of the invention herein described.